agent-bom discover gcp
Inventory GCP AI Assets to Generate a Standardized Bill of Materials
Install & Use
Copy this prompt and send it to your AI assistant (Claude / Cursor / TRAE / Codex / WorkBuddy etc.) to auto-install:
Help me install this AI Skill: agent-bom discover gcp. It is used for: Inventory GCP AI Assets to Generate a Standardized Bill of Materials Full Skill content: https://321skill.com/skills/agent-bom-discover-gcp-x-8/raw/index.md Read that page and install it.
The prompt includes a link to the full Skill content. You can also view the full content.
This Skill addresses the challenge of uniformly inventorying AI Agent and MCP-related assets on GCP. In practical operations, teams often need to manually log into the console or write ad-hoc scripts to query resources like Vertex AI, Cloud Run, Cloud Functions, and GKE, lacking a standardized inventory format, which leads to inefficiency in subsequent audits and compliance checks.
Usage is straightforward. You only need to ensure Python 3.11+ and the agent-bom library are installed locally, and configure GCP read-only credentials (ADC, Workload Identity, or short-lived service account). Then, by running a single command specifying the project and region, it automatically calls GCP APIs to discover all AI-related assets and outputs a JSON inventory file compliant with the agent-bom specification. The entire process does not require handing over long-term credentials to third parties; credentials always remain in the operator's environment.
It is ideal for operations teams, security/compliance personnel, and agent platform administrators who need to regularly audit GCP AI infrastructure. This is especially true for teams that have deployed multiple Vertex AI models, Cloud Run services, or GKE clusters and require a unified asset view for integration into CI/CD pipelines.
It is recommended to run this Skill after each significant change or periodically (e.g., weekly) to maintain an up-to-date asset inventory. Note that it performs discovery only and will not modify any GCP resources. The output path is specified by the operator, preventing data leakage.
Key Features
Unlike manually using gcloud commands or writing temporary scripts, agent-bom-discover-gcp outputs a standardized agent-bom JSON format that can be directly consumed by subsequent scanning tools. It also does not require exposing long-term GCP credentials to third-party services, offering higher security.
Limitations
Requires a Python 3.11+ environment, depends on the agent-bom library, and only supports GCP read-only credentials (ADC, Workload Identity, or short-lived service account). It does not support directly passing a service account JSON key file.
FAQ
Is it necessary to install the gcloud command-line tool?
Not strictly required, but installation is recommended for automatically reading default configurations. If using ADC or Workload Identity, gcloud can be omitted.
How do I specify the GCP project to inventory?
Specify it via the GOOGLE_CLOUD_PROJECT environment variable or the --project command-line argument. By default, it uses the project from ADC.
Installation guide for AI assistants
If your AI coding assistant (Claude Code, Cursor, TRAE etc.) can see this page, send it this message to auto-install:
Visit https://321skill.com/skills/agent-bom-discover-gcp-x-8/raw/index.md to read the original Skill definition (Markdown format) for agent-bom discover gcp, and install it according to the instructions.
Raw Markdown URL for AI: /skills/agent-bom-discover-gcp-x-8/raw/index.md